Open letters about AI development
Simon Willison · Simon Willison · 2026-08-02
Simon Willison summarizes three competing July 2026 open letters on AI policy: a Microsoft-led 235-company defense of open-weight models, Anthropic's separate push to crack down on large-scale distillation, and a 1,324-employee frontier-AI petition requesting government-paced automated AI development.
Appears in
Extraction
Topics: open-weight-modelsai-policyai-governancedistillationfrontier-ai
Claims
- A Microsoft-shepherded letter signed by 235 companies argues open-weight models reduce risk by distributing oversight and eliminating single points of failure.
- The open-weight letter explicitly endorses distillation as a legitimate model-development technique analogous to open-source software traditions.
- Anthropic CEO Dario Amodei published a separate response calling for a crackdown on industrial-scale distillation while denying Anthropic ever advocated a formal ban on open-weight models.
- A second letter signed by 1,324 frontier AI employees—including Dario Amodei, Ilya Sutskever, and Jakub Pachocki—requests U.S. government support for international efforts to deliberately pace automated AI research.
- Mounting evidence that AI companies are using AI to accelerate their own R&D (Anthropic generating 80% of code with Claude Code, OpenAI using Sol to cut serving costs 20%) is cited as motivation for the pacing letter.
Key quotes
Relying solely on closed models is not inherently safe: they can be breached, misused, or fail in ways that outsiders cannot detect. And concentrating advanced AI capabilities behind a small number of closed models compounds that risk.
Dario Amodei... called for 'a crack down on industrial-scale distillation operations', while also stating that 'Anthropic has never advocated for a ban on open-weights models'.
We request that the U.S. government support an international effort to develop the technical and governance tools needed to deliberately pace the frontier of automated AI development.